Mechanical properties of micro-alloyed steels studied using a evolutionary deep neural network

• Published in: Materials and manufacturing processes Vol. 35; no. 6; pp. 611 - 624
• Main Authors: Roy, Swagata, Saini, Bhupinder Singh, Chakrabarti, Debalay, Chakraborti, Nirupam
• Format: Journal Article
• Language: English
• Published: Taylor & Francis 25.04.2020
• Subjects: deep neural network; evolutionary computation; genetic programming; metamodeling; micro-alloyed steel; multi-objective optimization; neural net; parallel plot; reference vector; single variable response; Vanadium
• ISSN: 1042-6914; 1532-2475
• DOI: 10.1080/10426914.2019.1660786

The existing Evolutionary Neural Net Algorithm (EvoNN) for data-driven modeling has been augmented during this study using an evolutionary deep neural net strategy to give rise to a novel algorithm named EvoDN, which has been further upgraded to an improved version named EvoDN2. This study reports an application of EvoDN2 to study vanadium and niobium based micro-alloyed steels. For this purpose, a dataset for ultimate tensile strength, elongation and Charpy impact energy at −40°C is collected and trained using aforementioned EvoNN, EvoDN2, and another in house algorithm named Bi-objective genetic programming (BioGP). This trained models are then optimized to get optimized properties using a constrained version Reference Vector Evolutionary Algorithm (cRVEA). The results are thoroughly compared with the existing correlations and prior work and found to be well within the acceptable range.